I'm a new member to the coincommunity.com forum, but a longtime collector (40+ years) and pocket dealer (25 years).
The 1943-P 3/2 is one of the most accessible overdates* for collectors. PCGS estimates a survival pool of 17,000 in MS-60 and above, though that seems high. Given the high mintage for the 1943-P and the excessive die use for the date and mint, a large number of the 3/2 should be expected to exist in all grades. I certainly would not quibble with a survival rate of 20,000 to 25,000 in all grades. Over the years, I have accumulated just over two hundred junk silver rolls of the silver
War Nickels, mainly from roll lots and dealer junk boxes. Not surprisingly, the 1943-P dominates the bulk of those coins. In that time, I have located six of the overdates, with grades ranging from AG-3 to VF-35. Based on that experience, here are a few observations:
+ The 1943-P 3/2 is not rare. It isn't common, but it does exist in the wild and can be found by diligent searchers. It's s fun search.
+ Over the years, the most common question I have been asked about this particular overdate is "Is this one of them?" Only once has the proffered coin been an overdate. The key reason I think this coin generates the question is that people expect the overdate to be subtle, and hidden in the shadows. It isn't. The overdate is bold, and readily identifiable, even in my AG-3 example. Similar to the 1942 and 1942-D 2/1
Mercury dimes, the Philadelphia mint in this case simply made no effort to obliterate the "2" before rehubbing the die. This is understandable: it was wartime, speed mattered, quantity mattered, and the focus was on production.
+ Most of the confusion comes from the two die break varieties that show spikes protruding upward from the end of the lower curl of the "3." One variety has the crack proceeding vertically, pointed toward the top curl of the "3," while the other has a smaller crack pointing toward the middle bar of the "3." On a genuine 3/2, the spike protruding from the end of the bottom curl of the "3" points to approximately a 2:00 position within the bottom curl, a much more acute angle than the two die cracks.
+ Other confusion stems from the coin's composition. The copper-silver-manganese composition is prone to pitting, discoloration, metal flow lines, entrained bubbles, and laminations. These can mislead and obscure attribution, especially for people who haven't handled a lot of the coins. The soft composition also allowed for much more rapid wear, widening the gap between the oft-seen MS-65+ diagnostic pictures and the real world of grimy, worn G-to-VG coins.
+ Heavy die wear also contributes to the problem, with hairline die cracks, die scratches, clash marks and metal flow obscuring the key area.
All of that said, the overdate is quite bold. Once a collector has actually found an overdate, the search takes on a different dimension: the hairline cracks and scratches are quickly ignored, because the eye has been trained to look for the specific identifiers.
For me, the key diagnostic is the presence of the horizontal "check mark" or "fishhook." On coins of any grade, this remains bold. The horizontal line, in particular, lies in a well-protected area, surrounded by the lower curl of the "3." Even on my AG-3 specimen, this line is plainly visible, although the top of the "hook" or "check" starts to blend into the wear from the rim.
*I recognize the distinction between a older repunched overdates and a hubbed overdate. Thought the product of hubbing rather than repunching, the rehubbing here was done specifically to change the date. I think it is still fair to call this an overdate.